Re: Expressing metadata using IRW

Nathan, I think most of your message is off topic. Can we maybe
proceed in a somewhat more orderly fashion?  That is, can you answer
my specific question, how to express metadata using IRW, or some
comparably well-defined notational method?  I think it would be best
to start with the particular metadata example in my message, with the
solution being not idiosyncratic but as part of some general method.
Once we answer that, it will be easier to explore further.

For comparison, the httpRange-14 method for referring to the page is

  <http://lessig.org/blog/>

while the httpRange-14-independent method according to my "work in
progress" email would be something like

  [ wa:isServedFrom "http://lessig.org/blog/"^^xsd:anyURI ]

and Larry Masinter would have us write

  <duri:2011:http://lessig.org/blog/>

(well not exactly, as this refers to what the page was like in 2011,
which isn't the same thing... and the meaning of the duri: URI depends
on the meaning of the http: URI, which is muddied if we don't have
httpRange-14... but you get the idea)

and under the method where the referent is the retrieved page's
primary topic, we might create our own new web page describing Larry's
page somehow (probably including its URI; this begs the question), put
it somewhere such as http://example.com/z, and then refer to the page
using

  <http://example.com/z>

and with IRW one would write... ?

I was surprised that BIBO and CiTO don't seem to have ways to refer to
things on the web. BIBO has a class
http://purl.org/ontology/bibo/Webpage (oddly, a subclass of Document,
a "bounded physical representation of body of information") but no
relation connecting the Webpage to a URI.

Here are some other ideas that might be formalized if desired:
http://www.virtualsalt.com/mla.htm

In case it wasn't clear, solutions that require changes to the web
page are not very helpful, since in general the agent that wants to
refer to a page does not control its content. Even when Larry himself
uses the CC license chooser to decorate his page, he and the chooser
have very independent interests and competences, so there will be no
coordination between the content that they respectively contribute to
the document. In particular, the chooser has no idea what is on the
page or how the page's URI is otherwise being used.

With all due respect we have no shortage of personal opinions (and I
apologize for the many times I have given mine gratuitously). What we
need is an understanding of desiderata, current practice, and
potential practices, so that any AWWSW ontology might be capable of
accounting for them somehow, and so that others have tools that will
help them form their own opinions.

Best
Jonathan

Received on Tuesday, 25 January 2011 14:17:48 UTC